Abstract
Melt intercalation has typically been used to modify polymers with layered nanoparticles to obtain excellent performance. Polyphenylene sulfide (PPS)/graphene nanoplatelets (GNPs) composites were manufactured by one-step melt intercalation under shearing force. The morphology, thermal properties, and mechanical properties of the PPS/GNPs composites were then studied in detail. Graphene oxide showed poor thermal stability compared with GNPs. The GNPs were well dispersed in the PPS matrix, with a slight aggregation. Furthermore, the addition of GNPs accelerated the crystallization of PPS and reduced the incomplete crystallization of PPS due to the heterogeneous nucleation of GNPs. The thermal stability of the PPS/GNPs composite showed a significant improvement because of the good thermal conductivity and mass transport barrier effect of GNPs. The temperature corresponding to the heat-resisting index (T HRI) of the PPS/GNPs composites was also significantly improved due to the addition of GNPs. When the content of GNPs was only 0.5 wt%, the tensile strength and tensile modulus of the PPS/GNPs composites significantly increased compared with pure PPS; this could be attributed to the heterogeneous nucleation of nanoplatelets and the interfacial interaction between PPS and GNPs. The dynamic mechanical analysis of the PPS/GNPs composites indicated that the storage modulus of the PPS polymer increased significantly due to the addition of GNPs.
Keywords
Introduction
Since the first observation and preparation of graphene by Novoselov and Geim 1 in 2004, graphene nanoplatelets (GNPs) have attracted increasing interest in different fields due to their unique structure and excellent properties. Graphene is a two-dimensional monolayer of one-atom thickness that consists of sp 2-bonded carbon atoms arranged in a honeycomb structure. Due to its specific two-dimensional structure, graphene has extraordinary characteristics, such as high mechanical property (1 TPa Young’s modulus), excellent thermal conductivity (5000 W mK−1), a large surface area (2630 m2 g−1), the quantum Hall effect at room temperature, and good electrical properties. 2 –6 Therefore, graphene has varied potential applications, such as in electric devices, sensors, energy storage, super-capacitors, and biomedical applications. 7 –9 According to its structure, graphene can be roughly classified into graphene (one layer), bilayer graphene (double layer), few-layer graphene (3–5), and multilayer graphene (>5 layers and ≤10 nm). The preparation of single-layer GNPsis difficult and of low-efficiency, which limits their application in the field of composites. In comparison, the production of multilayer graphene is convenient and low cost. Furthermore, multilayer graphene has a high diameter/thickness ratio, excellent physical properties, and high thermal conductivity (3000–5000 W mK−1), which are very similar to the characteristics of single-layer graphene. 10 –12 Therefore, multilayer GNPs are considered to be effective nanofillers in the preparation of high-performance polymer composites.
Polyphenylene sulfide (PPS) is a high-performance engineering thermoplastic polymer constructed with a benzene ring linked by sulfur atoms. This material has high thermal stability, good mechanical properties, chemical resistance, fire resistance, and high dimensional stability due to its rigid structure. Because of its excellent performance, PPS has been widely applied in many fields, such as in heat sensor filters, high-temperature filters, automotive field, and aerospace industry. 13 –15 However, it should be noted that PPS is rather brittle, with a lower elongation at break. This feature limits the application of pure PPS; thus, commercial PPS is mostly modified by using polymers or nanoparticles.
Compared with pure polymers, graphene-based nanocomposites have excellent mechanical, electrical, thermal, and gas barrier properties even when the graphene content is very low. 16 –18 Thus, the addition of graphene may minimize the defects of PPS and result in new properties for the material. Some researches on PPS/graphene composites have been carried out in the last five years. The preparation of graphene-based composites can mainly be divided into three methods: in situ intercalative polymerization, solution intercalation, and melt intercalation. Zhang et al. 19 prepared PPS/exfoliated graphite (EGO) composites through solution intercalation and studied in detail the electric conductivity and mechanical properties of the manufactured PPS/EGO composites. The PPS resin was dissolved in 1-chloronaphthalene solution at 205°C. This preparation method was not regarded as an industrialization method of production. Zhao et al. 20 used a commercial exfoliated graphite to manufacture PPS composites by melt blending and then investigated the electrical conductivity of the PPS/EG composites. Gu et al. 21 used isopropyl trioleictitanate as a modifier to make surface modifications on GNPs. The PPS/GNPs nanocomposites were manufactured in two steps: the GNPs and PPS resin were first mechanically mixed in a ball mill machine at room temperature for 24 h and then were compression molded (295°C, 10 MPa) to produce the PPS/GNPs composites. The thermal percolation behavior of the PPS/GNPs composites was also studied in detail. The dispersion of GNPs in the polymer matrix has always been an obstacle in the preparation process. Graphene has poor dispersion and compatibility with organic polymer matrices because of the easy self-agglomeration of graphene layers due to the intrinsic van der Waals forces. 22 Functional graphene and graphene oxide (GO) usually have good compatibility with polymer matrices but show poor thermal stability at high temperature. However, the melt compounding temperature of PPS composites is about 300°C, which can lead to degradation of functional graphene or GO. Therefore, the direct use of GNPs to manufacture PPS/GNPs composites is hardly found in previous research. In contrast, graphite and EG have often been used to prepare PPS composites.
In the present work, PPS resin and GNPs are directly used to manufacture composites in a twin-screw extruder through a one-step process. In the melt blending, the rotation of the two screws can supply a shearing force that can be used to uniformly disperse GNPs. GO is also produced to evaluate its thermal stability compared with GNPs. Furthermore, the morphology, thermal, and mechanical properties of the obtained PPS/GNPs composites are investigated by scanning electron microscopy (SEM), X-ray diffraction (XRD), differential scanning calorimetry (DSC), thermogravimetric analysis (TGA), dynamic mechanical analysis (DMA), and tensile testing.
Experimental
Materials
PPS resin with a melt flow index of 150 g 10 min−1 (315°C 5 kg−1) was supplied by Jiangsu Ruitai Technology Co., Ltd (China). The GNPs sample, KNG-150, (1–20 μm in diameter and 5–50 nm thick), was obtained from Xiamen Knano Graphene Technology Co., Ltd (China).
Preparation of composites
GO was prepared in our research laboratory according to a previous method. 20 An SISZ-10A corotating twin-screw extruder (Wuhan RuiMing Co. Ltd., China) was used to manufacture the PPS/GNPs composites in a one-step process under shearing force. The loading fractions of GNPs in the PPS/GNPs composites were 0.5 wt%, 1 wt%, 3 wt %, and 5 wt%, respectively. The melt compounding lasted for 15 min at a screw speed of 30 r min−1 and at temperatures ranging from 270°C to 300°C. Both the PPS resin and GNPs were dried under vacuum at 100°C for 24 h before the compounding. The PPS/GNPs composites with various GNPs contents were labeled as PPSGNP x , where x is the content of GNPs. An SA-303 table-type test press machine (Tester Sanygo Co. Ltd, Japan) was used to obtain film specimens by hot pressing the banded composites at 295°C under 20 MPa. The film specimens were then cooled down to room temperature under 20 MPa. An SDL-100 sample cutting machine (Dumbbell Co. Ltd, Japan) was then used to incise the hot pressing film to obtain specimens for the testing of mechanical properties. Figure 1 shows the characteristics of the test specimens.

The shape and norms of test specimens for mechanical properties.
Characterization of the composite properties
The morphology of the PPS/GNPs composites was observed by using S-3000 N and SU1510 SEM (Hitachi High-Technologies Co. Ltd, Japan). The cryo-fracture surface of the composites was coated with a thin gold layer. The dispersion behavior of GNPs in the PPS matrix was also investigated by using a Miniflex 300 X-ray diffractometer (Rigaka Co. Ltd, Japan) with copper Kα radiation (λ = 0.154 nm) at 30 kV voltage and 10 mA current. The scattering angle ranged from 3° to 90° at 3°min−1.
A Q-200 differential scanning calorimetric device (TA Instruments Co. Ltd., New Castle, Delaware, USA) was used to measure the crystallization behavior of the PPS/GNPs composites. All DSC measurements were carried out in two steps under a nitrogen atmosphere (50 mL min−1). First, the samples were heated from room temperature to 320°C, at which they were kept for 3 min to eliminate the heat history. In the second step, the samples were cooled down to 30°C and then reheated up to 320°C at a heating and cooling rate of 20°C min−1. The thermal parameters for the crystallization and melting properties of the PPS/GNPs composites were obtained in the second test step. The crystallinity (X c) was calculated by applying the following equation:
where ΔH m is the melting enthalpy of completely melted composites, ΔH f is the melting enthalpy of 100% crystalline PPS, and W f is the mass fraction of GNPs content of the composites.
The thermal degradation behavior of the GO and PPS/GNPs composites was measured by using a Q-500 thermogravimetric analyzer (TA Instruments Co. Ltd) heated from 30°C to 800°C at 10°C min−1 under a nitrogen atmosphere (50 mL min−1).
The tensile properties of the PPS/GNPs composites were characterized by using an EZ-SX tensile tester (Shimadu Co. Ltd, Japan) with a 50-kN load cell at 25°C, with a strain speed of 5 mm min−1, according to JIS K7127. A DVA-225 dynamic mechanical analyzer (IT Keisoku Seigyo Co. Ltd, Japan) was used to investigate the dynamic mechanical properties of the composites. The film specimens were heated from room temperature to 300°C at a rate 10°C min−1 heating rate and a frequency of 10 Hz.
Results and discussion
TGA of GO and graphene
Figure 2 shows the TGA curves of GO and pristine graphene. Fourier transform infrared spectroscopy, Raman, and TEM analyses of GO have been previously done in our research laboratory. 23 The present work considers only the thermal stability of GO, which is found to be poor compared with pristine graphene. Graphene has negligible weight loss within the test temperature range, whereas GO has significant weight loss below 800°C, which can be attributed to the removal of physically absorbed water molecules and the degradation of oxygen-containing groups. Due to its obvious decomposition at the processing temperature of PPS (about 300°C), GO cannot be used as a nanofiller in the manufacture of PPS-based composites; its degradation may induce or accelerate the decomposition of PPS during the preparation process and thus influence the properties of the composites. The difference of thermal stability between graphene and GO can be attributed to the difference in their structure and chemical composition. Therefore, GO was not used to prepare PPS-based composites in this work.

TGA curves of graphene and GO. TGA: thermogravimetric analysis; GO: graphene oxide.
Morphology of the PPS/GNPs composites
The morphology of PPS and PPS/GNPs composites with various contents is investigated by SEM, as shown in Figure 3. The addition of GNPs clearly has a significant impact on the morphology of the composites. The fracture surface of the composites becomes rough with increasing GNPs content compared with pure PPS resin. Further, GNPs particles are found to be well dispersed in the PPS matrix. However, it should be noted that GNPs can form a slight agglomeration on the fracture surface of composites with a diameter smaller than 2 μm.

SEM images of PPS (a) (e), PPS/GNP1 (b) (f), PPS/GNP3 (c) (g) and PPS/GNP5 (d) (h) composites. SEM: scanning electron microscopy; PPS: polyphenylene sulfide; GNP: graphene nanoplatelets.
The dispersion and intercalated structure of GNPs in the PPS matrix are also characterized by XRD. As shown in Figure 4, a new peak at 2θ = 26.47° appears in the XRD patterns of the PPS/GNPs composites, and this peak becomes much sharper with increasing GNPs content. According to the literature, 24 this sharp peak can be attributed to the diffraction peak (002) of expanded graphite. It can be inferred that some GNPs self-aggregate during the melt blending due to the van der Waals forces. 22 Hence, obtaining a much better intercalation or exfoliation structure by simple melt blending under shearing force presents a challenge.

XRD patterns of PPS and PPS/GNPs composites. XRD: X-ray diffraction; PPS: polyphenylene sulfide; GNP: graphene nanoplatelets.
Crystallization behavior of the composites
Figure 5 shows the DSC curves of the heating and cooling process of the PPS and PPS/GNPs composites. Table 1 presents the thermal parameters of the PPS and PPS/GNPs composites, as summarized from the DSC curves. As shown in Figure 5(a) and Table 1, the addition of GNPs can increase the onset temperature of melting (T mo), the peak melting temperature (T m), and the melting enthalpy (ΔH m). This indicates that the crystalline perfection of PPS increases with the addition of GNPs. The data in Figure 5(b) and Table 1 show a significant change in the onset temperature of crystallization (T co) and the peak crystallization temperature (T c) of PPS, when GNPs are used to modify PPS. The T c and T co of the PPS composites first increase with the GNPs content but later gradually decrease. When the content of GNPs is 3 wt %, the T c and T co of the PPS/GNPs composites are about 9°C and 11°C higher, respectively, than those of pure PPS resin.

DSC curves of PPS and PPS/GNPs composites: (a) reheating curves and (b) cooling curves. DSC: differential scanning calorimetry; PPS: polyphenylene sulfide; GNP: graphene nanoplatelets.
DSC parameters of PPS and PPS/GNPs composites.
PPS: polyphenylene sulfide; GNP: graphene nanoplatelets.
Generally, the effects of GNPs on the crystallization behavior of PPS can be classified into two types. First, the addition of GNPs can promote the crystallization of PPS as a result of the heterogeneous nucleation of GNPs. Second, GNPs can hinder the mobility of PPS chains and inhibit the crystallization by increasing the viscosity of the composites. 20,25 The degree of supercooling (ΔT = T m − T c) required for the crystallization of PPS in PPS/GNP3 composites is lower by about 7°C than that required for pure PPS. This decreased degree of supercooling suggests an accelerated crystallization rate of PPS due to the presence of GNPs. Therefore, when the content of GNPs is less than 3 wt%, PPS composites have a faster crystallization rate and higher crystallinity than pure PPS. This first type of effect is dominant. However, when the content of GNPs increases to 5 wt%, the crystallization rate and crystallinity decrease slightly but remain higher compared with pure PPS. This phenomenon can be attributed to the restricted motion of PPS chains or to a mass barrier effect due to the presence of GNPs. This second type of effect is more gradual.
Thermal stability of the composites
Figure 6 shows the TGA and DTG curves of the PPS and PPS/GNPs composites. As indicated in the figure, both pure PPS resin and PPS/GNPs composites undergo a single-stage thermal degradation process over the whole test temperature range. Based on the data in Table 2, PPS starts to decompose at around 450°C, whereas PPS/GNPs composites degrade at about 30°C higher temperature compared with pure PPS. The thermal stability of PPS clearly shows a significant increase with the addition of GNPs. Meanwhile, the temperature corresponding to weight losses of 15 wt% (T 15%) and 50 wt% (T 50%) for neat PPS resin are 485°C and 534°C, respectively. The T 15% and T 50% for PPS composites also increase by around 16–22°C and 17–39°C, respectively. Furthermore, the T max of the composites increase by around 9–20°C due to the addition of GNPs. The improvement in thermal stability can be attributed to the mass transport barrier effect of GNPs, which restricts the emission of volatile degradation products during the decomposition, thus slowing down the degradation. Such improvement may also be due to the good thermal conductivity of GNPs, which can facilitate the heat dissipation within the PPS composites. Therefore, a higher temperature and a longer time are required for PPS composites to reach the initial decomposition temperature. 19

TGA (a) and DTG (b) thermograms of PPS and PPS/GNPs composites. TGA: thermogravimetric analysis; DTG: differential thermogravimetry; PPS: polyphenylene sulfide; GNP: graphene nanoplatelets.
TGA parameters of PPS and PPS/GNPs composites.a
PPS: polyphenylene sulfide; GNP: graphene nanoplatelets; TGA: thermogravimetric analysis.
a T HRI = 0.49 × [T 5 + 0.6 × (T 30 − T 5), where T 5 and T 30 is the corresponding decomposition temperature of 5% and 30% weight loss respectively.
The heat-resistant index (T HRI) can be used to express the thermal stability of PPS composites, which can represent the limiting temperature of long time serves. As shown in Table 2, the T HRI values of the PPS/GNPs composites first increase and then decrease with increasing GNPs fillers. This phenomenon indicates that the thermal stability of the PPS/GNPs composites also first increase and then decrease with increasing of GNPs content. This can be attributed to the higher heat capacity and thermal conductivity of GNPs compared with pure PPS, allowing the nanoplatelets to more easily absorb external thermal energy. Moreover, the favorable dispersion of GNPs in the PPS matrix can enhance the thermal stability of PPS/GNPs composites. However, when the content of GNP fillers is high, GNPs agglomerate in the PPS matrix, which can decrease the relative specific heat capacity of GNPs and damage the overall structure of PPS composites. Therefore, the thermal stability of PPS composites decreases with increasing content of GNPs. 26 –28
The difference in temperature (T max − T 5%) can usually be regarded as a measure of the degradation rate. It should be noted that the difference in temperature is about 60°C for pure PPS but only about 50°C for PPS/GNPs composites. This indicates that composites have a faster degradation rate compared with pure PPS. Furthermore, the thermal stability of PPS composites also slightly decreases with increasing GNPs content but remains better than that of pure PPS. All these phenomena may also be explained by the high thermal conductivity of GNPs. The residual amount of PPS/GNPs composites is also found to be much higher than that of pure PPS resin due to the chemical interactions between the degradation products of PPS and GNPs.
Mechanical properties of the composites
Figures 7 and 8 show the tensile properties and stress–strain curves of the PPS and PPS/GNPs composites, respectively. The addition of GNPs is found to have a significant influence on the tensile strength and tensile modulus of the composites, both of which first increase with the GNPs content in general, and then gradually decrease. When the GNPs content is only 0.5 wt%, the tensile strength and tensile modulus increase from 50 MPa and 1174 MPa to 113 MPa and 2704 MPa, respectively. Therefore, GNPs have a significant enhancing effect on the tensile properties of PPS. The addition of GNPs can act as a heterogeneous nucleation agent that improves the crystallization and increases the crystallinity of the PPS matrix, which can enhance the mechanical property. Moreover, the appropriate addition of GNPs (≤1 wt%) can effectively transfer the external load and prevent crack propagation due to the interfacial interaction between the GNPs and PPS matrix, 10 –12,29 which can improve the tensile property. However, when the content of GNPs is 3 wt%, the tensile strength and tensile modulus decrease compared with pure PPS. This can be explained by the fact that GNPs can easily agglomerate in the PPS matrix with the increase in GNPs content, resulting in more interfacial defects and stress concentration points, which can decrease the mechanical property of PPS/GNPs composites. 10 –12 However, it should be noted that when the GNPs content reaches 5 wt%, the tensile properties of the composites show an unexpected improvement. We believe that, with a higher content of GNPs fillers, some GNPs can agglomerate in the PPS matrix, which decreases the tensile properties. However, the nanolayer of well-dispersed GNPs may be sufficient to form the network structure, which can significantly increase the mechanical properties, thus the unexpected increase in the tensile properties of the PPS/GNPs composites. As shown in Figure 8, both PPS and PPS/GNPs composites have the characteristic of brittleness, the elongation at break of the PPS/GNPs composites is slightly higher than that of pure PPS. Furthermore, the elongation at break of the PPS/GNPs composites (except PPS/GNP5) decreases with increasing the GNPs content.

Tensile properties of PPS and PPS/GNPs composites: (σ b) tensile strength and (E) tensile modulus. PPS: polyphenylene sulfide; GNP: graphene nanoplatelets.

The stress–strain curves of PPS and PPS/GNPs composites. PPS: polyphenylene sulfide; GNP: graphene nanoplatelets.
DMA can be used to confirm the performance of composites under stress and at various temperatures. Figure 9 shows the variation in the storage modulus of PPS and PPS/GNPs composites according to the temperature. The PPS/GNPs composites show higher storage modulus values over the whole test temperature range, especially below the glass transition temperature, compared with pure PPS. The storage modulus of the composites also increases with increasing GNPs content, which can be attributed to the reinforcing effect of GNPs in PPS/GNPs composites. Furthermore, the storage modulus of both pure PPS and PPS/GNPs composites decreases as the temperature increases. This phenomenon can be attributed to the increase in the molecular mobility of PPS polymer chains due to the increase in temperature.

The storage modulus versus temperature of PPS and PPS/GNPs composites. PPS: polyphenylene sulfide; GNP: graphene nanoplatelets.
Figure 10 shows the tan δ peaks of pure PPS and PPS/GNPs composites. In fact, the intensity of tan δ at T g can be regarded as an indicator of the energy-damping characteristic of the polymer and also implies the cooperative nature of the relaxation process of polymer chains. As shown in Figure 10, the peaks of tan δ move to a slightly lower temperature with increasing GNPs content. Moreover, the value of tan δ also decreases with the increase in GNPs content. This indicates that the hindering effect of GNPs on PPS chains and PPS composites weaken at dissipating energy levels. The reason for this may be that the incorporation of GNPs in the PPS matrix can introduce interfacial voids to melt blending composites, which enhances the free volume of the composites to a certain extent and may lead to a decrease in T g.

The tan δ versus temperature of PPS and PPS/GNPs composites. PPS: polyphenylene sulfide; GNP: graphene nanoplatelets.
Conclusion
PPS/GNPs composites were prepared by direct melt intercalation under shearing force. GO was not suitable for the preparation of composites by melt blending due to its poor thermal stability compared with graphene. The morphology of the composites was characterized by using SEM and XRD, which showed that GNPs dispersed uniformly in the PPS matrix, with a slight agglomeration due to the van der Waals forces. The DSC results indicated that the addition of GNPs could accelerate the crystallization rate of PPS and increase the crystal perfection of PPS due to the heterogeneous nucleation of GNPs. However, when the GNPs content reached 5 wt%, the GNPs would hinder the mobility of PPS chains and inhibit the crystallization. Furthermore, the addition of GNPs could significantly improve the thermal stability of PPS resin but increase the degradation rate of PPS composites. The T HRI also increased due to the addition of GNPs. When the content of GNPs was only 0.5 wt%, the tensile strength and tensile modulus of the composites significantly increased, whereas the elongation at break increased only slightly, which could be attributed to the heterogeneous nucleation of nanoplatelets and the interfacial interaction between PPS and GNPs. The addition of GNPs could significantly increase the storage modulus of composites, and the peaks of tan δ were found to shift to a slightly lower temperature. Furthermore, the value of tan δ also decreased with the increase in GNPs content.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The support of this research was supplied by the Fundamental Research Funds for the Doctoral Graduate of Jiangnan University (no. JUDCF13024) and Scientific Research Innovation of Ordinary Colleges of Jiangsu Province (no. CXZZ13_0748).
